Squeak-dev September 2001

squeak-dev@lists.squeakfoundation.org

168 participants
408 discussions

Re: CRLF problem with Squeak3.1a-4332.zip on FTP server?
by Bernhard Pieber 28 Sep '01

28 Sep '01

Bruce, Henrik, Ned, Thank you very much for looking at my problem and for the prompt solution! I have just downloaded Squeak3.1a-4347b.zip and the problem is gone. Now I managed to install Zurgle which this corrupt method prevented in the first place. Thanks again. Bernhard

1 0

Apple asking/lisiening
by Jeff Szuhay 28 Sep '01

28 Sep '01

here's what I received today from John Martellaro in Apple's Science and Technology Group (you can join this mailing list at http://lists.apple.com) here's it is >Date: Fri, 28 Sep 2001 10:14:04 -0600 >Subject: ASAG question # 15 ** Important ** > >DEADLINE: October 3rd. > >Okay, folks. This is the _big_ one. I have been asked to >query you on the top Unix development tools (NOT libraries, >that was Question # 12) that you want to see brought to Mac OS X. Well, even though I know some are working on getting Squeak on OS X, wouldn't it be great to have a bit of attention from Apple? The more people ask for it, the more likely it'll be done. Jeff Sz. -- ************************************************** Jeff Szuhay A randomly-directed www.szuhay.org chaotical wetware pattern jeff(a)szuhay.org recognizer/generator. If at first the idea is not absurd, then there is no hope for it. -- Albert Einstein

1 0

[ENH?] BrowserSelfReferences
by danielv＠netvision.net.il 28 Sep '01

28 Sep '01

from preamble: "Change Set: BrowserSelfReferences Date: 28 September 2001 Author: Daniel Vainsencher Browser has explicit references to itself, breaking subclasses. Replace them with self. Could be this change breaks things. Anyone happen to know?"!

1 0

Re: CRLF problem with Squeak3.1a-4332.zip on FTP server?
by Bruce ONeel 28 Sep '01

28 Sep '01

Thanks very very much! The new .zip is up on the ftp site in the usual place, ie Squeak/3.0/unstable-testPilot.. cheers bruce Henrik Gedenryd <Henrik.Gedenryd(a)lucs.lu.se> wrote: > Bernhard Pieber wrote: > > > It seems that my first assumption was false. I opened a file list on the > > extracted changes file Squeak3.1a-4332.changes. I viewed it as hex and > > searched for 0A, which should be the linefeed character. There is none in > > there. So, it is not the CRLF problem. However, ArchiveViewer>>extractAll is > > still corrupt. The code pane in the browser shows the following source: > > 61.image----#(19 August 2001 12:51:39 pm) > > > > What can it be? > > > > Thanks for your help. > > Bernhard, > > Somehow a glitch crept into the source bindings when I created the updates. > We discovered one such problem back then, but obviously one persisted. I > don't know what may have caused it. Recompiling all classes, there were > problems in ArchiveViewer and Debugger. I still don't know what causes this > problem but it does seem to happen every now and then. > > I have now created a new Squeak3.1a-4347b.zip from the 3.0 image (loading > all 800 change sets!), and I'm asking Bruce to remove all old zips and > replace them with this new one, where I've verified that all methods' > sources are ok. > > All previous packages at 4261 and up have had this glitch. If this presents > a problem, then switch to the new zip package, to get rid of this bad > method. > > Sorry for the inconvenience, > > Henrik

1 0

RE: [rant] Re: [OT] Will the SSSCA outlaw Squeak?
by Richard A. O'Keefe 28 Sep '01

28 Sep '01

Stephen Pair wrote: Cees, if you think that I, as an American, am going to stop for even a second and ponder why this act was committed, you are sorely mistaken. My only concern in this matter is justice. This is getting us even further off topic, but HOW THE HELL CAN WE HAVE JUSTICE WITHOUT PONDERING? It is *essential* to understand why the act was committed in order to take *effective* steps to stop further deaths of the innocent. It is precisely the refusal to ponder on *both* sides which has brought us this far. I fully agree that the people who did this dreadful thing must be brought to justice. But what should be done to stop other people wanting to do the same kind of thing? Wars have a terrible habit of killing even more innocent people, and laying up hatred for more wars in the future. (Like WWI leading to WWII.) That isn't justice. If I became dictator of this country, I would do two things: 1. Cause to be written in large letters in the House of Parliament and in every court this quotation from Oliver Cromwell: Brethren, I beseech you in the bowels of Christ, CONSIDER IT POSSIBLE THAT YOU MAY BE WRONG. 2. Resign.

19 28

[Ann] Codecrawler analysing Squeak
by Stephane Ducasse 28 Sep '01

28 Sep '01

Hi we made public the visualWorks image containing squeak3.0 and codecrawler so that if you get the vm of VW you can look at Squeak with another point of view ;) Look at the page: http://www.iam.unibe.ch/~lanza/CodeCrawler/SqueakInCC/codecrawlerSqueak.html Look especially at the blueprints.... Have fun.

1 0

help with Socket>>isOtherEndConnected
by Michael Fremont 28 Sep '01

28 Sep '01

Hi all, I'm a squeak newbie esp. wrt sockets. I have a telnet connection open to squeak and run this code on the connected socket: (Delay forSeconds:10) wait. socket isOtherEndConnected ifTrue: socket sendData: 'hi there'. This works, including if I close the other end before I get to this code (no send is then attempted). But the following code gets a debug message, 'send data timeout; data not sent' when I close the other end well before the test: 10 timesRepeat: [ (Delay forSeconds:10)wait. socket isOtherEndConnected ifTrue: socket sendData: 'hi there'. ] I'm going bonkers trying to figure this out. With tests of various permutations the failure seems to happen anytime the test is inside a block. I am running RedHat Linux 7.1, squeak vm 3.0pre2 and image updated to #4343. I get the same result with the latest vm compiled from Lex Spoon's patches page. Any ideas? Thanks in advance, Mike

2 2

RE: [Ann] prealpha port of Ginsu available at Squeak Foundation
by Peter Crowther 28 Sep '01

28 Sep '01

> From: Hans-Martin Mosner [mailto:hmm@heeg.de] > So tonight I'll try to play with Ginsu and see how I can make > minced meat of my image. Ah yes, the old 'power tools' problem that I warn my delegates about. Used with understanding and proper training, they allow you to work much more efficiently than before. Used without understanding and proper training, they allow you to cut your own legs off much more efficiently than before... "Smalltalk: The ultimate power tool. It slices, it dices, it carves into little tiny pieces... anything, including your image. Get one today." - Peter

1 0

Re: [modules] documentation pages on the swiki
by danielv＠netvision.net.il 28 Sep '01

28 Sep '01

A little note about how we're going about it - In the spirit of modularizing the modularity effort, I'm not sure there's a need to centrally cut everything up. I propose this way of doing things - * Get the Modules code in the image, * connect things to a repository, * publish a small step by step recipe to carving out a proper module. and then see my recent Updates mail. Given code and something clear to do with it, I think people will take care of the rest quite naturally. I think it's better if he who will maintain a module marks his own "territory". Just a thought. Daniel Henrik Gedenryd <Henrik.Gedenryd(a)lucs.lu.se> wrote: > I have now created a page on the swiki for building documentation about the > modules system. Currently it lists the specification that exists, plus > copies of mail discussions that may be helpful. > > http://minnow.cc.gatech.edu/squeak/2042 > > (Unfortunately writing "modules" instead of 2042 doesn't work since the > swiki gives the first match not the best match. And while I'm complaining, > the swiki has been practically unusably slow, since midnight at least, my > time.) > > In particular, there is a page with the proposal for an image architecture > that I mentioned in my reply to Stephane. > > http://minnow.cc.gatech.edu/squeak/2057 > > If people agree with this, it would become the basis for refactoring the > image into a modular Italian dish. Ah, what the, I'll copy in the text here: > > A proposed image architecture > General principles for reorganizing the existing image into separable > modules. > > > Goals for the module structure > There is so much reorg. needed for "doing it right" that we need to decide > on a few reasonable goals that are our most immediate needs and focus on how > to solve those first. > > I propose these: An ability to have the equivalent of a majorShrink image. > Also a PDA image with just a basic Morphic; this mostly involves separating > eToy from Morphic--Balloon, 3D, etc. are easier. More, a completely headless > image (very small, neither MVC nor Morphic). This is a secondary goal. > > With these goals in mind, I have started to analyze the code, and I wrote > some tentative reorganization code. I realized that we need to decide on a > modular architecture for reorganizing the system. That is, a few overall > principles for how the modular system should be organized. Then people could > adopt these, take on the image part of their choice, and start working on > refactorings and submitting the code. > > > An "onion skins" module architecture > Wrt such principles, I saw a repeating pattern. Most "major" category > functionality (Morphic, Network...) can be divided into these "minor" > categories: Essential, Useful, Applications, Tests, and > Demo/documentation/examples. This pattern works for Morphic, 3D, Network, > FFI, Archives, Exceptions, Kernel/System, and more. It is like the "onion > layered" OS architecture, where further in is more essential, and modules > may use other modules in layers further in, but not further out. It also > provides useful segments for building different images for various purposes. > > > * Core - Essential parts that can1t be removed or split, like the core of > Morphic. > * Library - reusable utility parts that are much like Core but > optional/removable, such as various, generally useful morph classes > * Applications - larger, more standalone/2finished2 pieces. examples: > Morphic PDA, B3D Alice/Wonderlands. Network: Scamper, Celeste, etc. > * Tests - This is for SUnits > * Demo - Demo and example code. nNote that example code often contains a > lot of the degenerate cross-modular references. (This is also true for tests > and applications.) > > Note that we'll need to rename "Kernel", it is so similar to Core to be > confusing. Also "System" is a very weird major category--what system? > Squeak? The OS? etc. My idea was to instead have Language and Technology > (vs. "Media"--3D, speech, sound, etc.), where these needn't correspond > directly to Kernel and System. > > Moreover, "Tools" is highly ambiguous too. Perhaps put these under a new > "Development" header, to hold everything that could be purged from a > finished application: browsers, debugging tools, compiler, etc.

1 0

[ANN][RB] Re: One last alpha version of the RB...
by danielv＠netvision.net.il 28 Sep '01

28 Sep '01

Hi, everyone. I've just published a new version of RB, integrating Bob's latest. I also added all the various files that lived on Bob's download directory, specifically the Lint UI and the Model tests. This is the SCAN page - http://squeak.heeg.de:8080/TFODC69UQ0MFRIHMG40DRML1CU7VYZ9 and to load it all directly from Squeak (4332), do - Project fromUrl: 'http://squeak.heeg.de:8080/TFODC69UQ0MFRIHMG40DRML1CU7VYZ9-3/Squeak RB.pr.gz' About the SF/SCAN issue - we'll see about mirroring the SCAN site on SF, and thus get the best of both worlds. Have well factored day! :-) Daniel Bob Hartwig <bob(a)bobjectsinc.com> wrote: > This version does NOT have > Daniel et. al.'s Browser integration work included, but it does have a > RBModel changeset that greenlights under Squeak 3.1 alpha. ... > Several Squeakers have taken the opportunity of CS3 to further > the browser integration work, and they have volunteered to take over future > integration work. You can find the fruits of their labors <a > href="http://squeak.heeg.de:8080/TFODC69UQ0MFRIHMG40DRML1CU7VYZ9">Here</a>. > -Bob

12 16

← Newer
1
...
4
5
6
7
8
9
10
...
41
Older →

Jump to page: